Patrice Lopez

Simple vs. sophisticated approaches for patent prior-art search

Patent prior-art search is concerned with finding all filed patents relevant to a given patent application. We report a comparison between two search approaches representing the state-of-the-art in patent prior-art search. The first approach uses simple and straightforward information retrieval (IR) techniques, while the second uses much more sophisticated techniques which try to model the steps taken by a patent examiner in patent search. Experiments show that the retrieval effectiveness using both techniques is statistically indistinguishable when patent applications contain some initial citations. However, the advanced search technique is statistically better when no initial citations are provided. Our findings suggest that less time and effort can be exerted by applying simple IR approaches when initial citations are provided.

PATATRAS: retrieval model combination and regression models for prior art search

This paper presents PATATRAS (PATent and Article Tracking, Retrieval and AnalysiS), a system realized at the Humboldt University for the IP track of CLEF 2009. Our approach presents three main characteristics: 1. The usage of multiple retrieval models and term index definitions for the three languages considered in the present track producing ten different sets of ranked results. 2. The merging of the different results based on multiple regression models using an additional training set created from the patent collection. 3. The exploitation of patent metadata and the citation structures for creating restricted initial working sets of patents and for producing a final re-ranking regression model. The resulting architecture allowed us to exploit efficiently specific information of patent documents while remaining generic and easy to extend.

Repair strategies for lexicalized tree grammars

This paper presents a framework for the definition of monotonic repair rules on chart items and Lexicalized Tree Grammars. We exploit island representations and a new level of granularity for the linearization of a tree called connected routes. It allows to take into account the topology of the tree in order to trigger additional rules. These local rules cover ellipsis and common extra-grammatical phenomena such as self-repairs. First results with a spoken language corpora are presented.